**EU-backed Coalition Aims to Shape Global AI Governance**
A coalition of mostly European nations and middle powers has emerged with the intent of establishing a new global framework for artificial intelligence (AI), seeking to influence the governance of rapidly advancing technologies. This initiative, which includes leaders from 20 countries, was formally announced on Monday during the UN General Assembly High-level Week in New York. European Commission President Ursula von der Leyen was among those who signed the declaration, which emphasizes the necessity for human oversight in the development and deployment of frontier AI systems.
The coalition, spearheaded by Finnish President Alexander Stubb and Norwegian Prime Minister Jonas Gahr Store, aims to ensure that AI technologies remain "under human direction, oversight and control." The declaration highlights the potential benefits of AI, such as improving quality of life, advancing scientific research, and bolstering economies. However, it also acknowledges the risks associated with the rapid evolution of AI systems, including their ability to bypass safety measures and exploit vulnerabilities in real-world applications.
Notably absent from the signing were the United States and China, the two leading powers in AI development. Several other major European nations, including the United Kingdom, France, and Italy, also did not participate. However, Germany, Spain, and a number of smaller EU states lent their support to the initiative. The declaration remains open for additional endorsements from other countries.
The coalition's statement calls for several key measures to enhance AI governance. These include mandatory pre-deployment testing, independent evaluations with adequate access to advanced AI models, the establishment of common international standards, and the shared reporting of significant safety incidents. Importantly, the declaration does not advocate for a halt in AI development but suggests that UN member states consider the creation of an international body tasked with setting standards, ensuring compliance, and convening governments when AI systems reach certain capability thresholds.
Stubb emphasized the importance of involving the dominant AI powers, stating that the coalition's goal is to create "momentum" that could eventually attract both the US and China to the table. He argued that it is in the best interest of these nations to establish "guardrails" for AI development, asserting that human oversight is essential. Stubb proposed a model akin to the International Atomic Energy Agency, although the coalition has intentionally left the institutional structure open to avoid alienating governments that may be hesitant to create another UN body.
The formation of this coalition comes at a time when the US and China are increasingly defining the global AI landscape and promoting competing frameworks for AI governance. The US has expanded its Pax Silica initiative, which focuses on securing semiconductor and AI supply chains among allied nations, now involving 24 partners. In contrast, China has launched the World Artificial Intelligence Cooperation Organization (WAICO), which includes 29 founding countries and aims to position itself as a leader in AI development, particularly in the Global South.
The UN has also initiated its own Global Dialogue on AI Governance, leading to a competitive environment for establishing global AI regulations. Some countries are adopting a multifaceted approach to AI governance, engaging with multiple emerging frameworks. For instance, Kazakhstan has joined both the WAICO and the US-led Pax Silica initiative while also endorsing the European-led declaration. Similarly, Kenya and South Africa, founding members of WAICO, have expressed support for the new coalition.
As the global discourse surrounding AI governance evolves, the newly formed coalition seeks to carve out a significant role in shaping the future of AI regulation, advocating for responsible development that prioritizes human oversight and international cooperation. The coming months will likely reveal how effectively this coalition can influence the broader AI governance landscape amid the competing interests of major global powers.